Semantic Fusion for Connected Medicine: An Experimental Comparison of Tabular and Ontological Structures in the Efficient Management of Diabetes

Blaise Muhala; Munduku Munduku Deo; John Bukasa Kakamba; Oshasha Oshasha Fiston; Rostin Mabela Makengo; Celestin Muluba; Djonive Munene

doi:10.14232/analecta.2025.2.46-66

Authors

Blaise Muhala University of Kinshasa
Munduku Munduku Deo University of Kinshasa
John Bukasa Kakamba University of Kinshasa https://orcid.org/0000-0003-1031-2326
Oshasha Oshasha Fiston University of Kinshasa https://orcid.org/0009-0009-5447-2760
Rostin Mabela Makengo University of Kinshasa
Celestin Muluba University of Kinshasa
Djonive Munene University of Kinshasa https://orcid.org/0009-0006-1640-9216

DOI:

https://doi.org/10.14232/analecta.2025.2.46-66

Keywords:

medical ontology, intelligent assistant, knowledge representation, connected medicine

Abstract

The explosion in digital medical data makes it crucial to design intelligent assistants capable of interrogating this information reliably, quickly and contextually. In this article, we propose a novel comparative approach between two forms of knowledge representation: traditional tabular data and semantic ontologies. Based on the same clinical dataset concerning diabetic patients, we have implemented a dual structuring: a tabular version and an RDF ontology modelled with Protégé. An intelligent assistant, interfaced with the GPT-4 API, was designed to query both formats. The originality of our contribution lies in the experimental parallelisation of these two data models, through a standardised series of 300 questions, classified according to three levels of increasing complexity. This methodology enables us to objectively assess the robustness, responsiveness and inference capacity of each approach. The results are unequivocal: the ontology systematically outperforms the tabular format, with exact response rates ranging from 97% to 100%, compared with 34% to 81% for the tabular format. In addition, the ontological approach shows better tolerance of ambiguous queries and stability in semantic interpretation. Over and above performance, this study highlights the potential of knowledge graphs as an architectural foundation for future medical decision support systems. It also paves the way for hybrid systems that combine the accessibility of tables with the semantic power of ontologies - a perspective that has so far been little explored in the context of connected healthcare.

Downloads

Download data is not yet available.

References

[1]. Bin.Z., Archie.W., Edward.G., Worldwide trends in diabetes prevalence and treatment from 1990 to 2022: a pooled analysis of 1108 population-representative studies with 141 million participants, in The Lancet, Nov. 2024, vol. 404, no. 10467, pp. 2077–2093

[2]. American Diabetes Association, Diagnosis and Classification of Diabetes Mellitus, Diabetes Care, vol. 32, suppl. 1, pp. S81–90, 2014

[3]. Rukundo T. B., Prevalence and Epidemiology of Diabetes in West Africa: Trends and Challenges, Research Invention Journal of Biological and Applied Sciences, décembre 2024, vol. 2,n°4, pp. 67–74.

[4]. Gakidou E., Afshin A., Abate D., et al., Global, regional, and national burden of diabetes from 1990 to 2021, with projections to 2050: a systematic analysis for the Global Burden of Disease Study 2021, The Lancet, juin 2023, vol. 402, n°10397, pp. 203–234.

[5]. Ejiofor Ugwu, Samuel Onung, Ignatius Ezeani, Michael Olamoyegun, Olufunmilayo Adeleye, Andrew Uloko, Barrières aux soins du diabète dans un pays en développement : preuves exploratoires des prestataires de soins, in Journal of Advances in Medicine and Medical Research, juillet 2020, vol. 32, n°10, pp. 72–83.

[6]. Topol, Eric J. « High-Performance Medicine: The Convergence of Human and Artificial Intelligence ». Nature Medicine, vol. 25, no 1, janvier 2019, p. 44‑56..

[7]. Esteva A., Robicquet A., Ramsundar B., et al., A guide to deep learning in healthcare, dans Nature Reviews Drug Discovery, juin 2019, vol. 21, n°6, pp. 395–415. Disponible sur :

[8].Karl W. Broman, Kara H. Woo, Data Organization in Spreadsheets, in The American Statistician, vol. 72, n°1, pp. 2–10, 2018

[9]. Palojoki S., Lehtonen L., Vuokko R. Semantic Interoperability of Electronic Health Records: Systematic Literature Review, dans JMIR Medical Informatics, vol. 12, n°1, janvier 2024

[10]. Altuhaifa F., Al Tuhaifa D., Developing an Ontology Representing Fall Risk Management Domain Knowledge, dans Journal of Medical Systems, avril 2024, vol. 48, article 47.

[11]. Noy N.F., McGuinness D.L., Ontology development 101: A guide to creating your first ontology, dans Applied Ontology, 2022, vol. 17, n°1, pp. 65–85. Disponible sur :

[12]. Antoniou G.,van Harmelen F.,A Semantic Web Primer,MIT Press,3eédition,2023,chap.5,pp.101–120.

[13]. Ahmed Shihab Ahmed, Safa Bhar Layeb, Systematic Review of Web-Based Decision Support Systems for Clinical Applications: Enhancing Ontology with Unified Modeling Language and Ontology Web Language, in TEM Journal, février 2024, vol. 13, n°1, pp. 77–89.

[14]. P.Hager,F. Jungmann,R. Holland. Evaluation and Mitigation of the Limitations of Large Language Models in Clinical Decision Making, dans Nature Medicine,vol.30,pp.2613–2622, juillet 2024

[15]. Lee J., Yoon W., Kim S., et al., BioBERT: A pre-trained biomedical language model for biomedical text mining, dans Bioinformatics, février 2022, vol. 38, n°2, pp. 1234–1240.

[16]. Karan Singhal,Tao Tu,Juraj Gottweis, Rory Sayres, Ellery Wulczyn, Mohamed Amin, Le Hou, Kevin Clark,Stephen R.Pfohl,Heather Cole-Lewis,Darlene Neal,Qazi Mamunur Rashid, Mike Schaekermann, Amy Wang, Dev Dash, Jonathan H. Chen, Nigam H. Shah, Toward expert-level medical question answering with large language models, in Nature Medicine, Janvier 2025, vol. 31, n°3, pp. 943–950

[17]. Zhen Hou, Hao Liu, Jiang Bian, Xing He, Yan Zhuang., Enhancing Medical Coding Efficiency Through Domain-Specific Fine-Tuning of Large Language Models, dans Nature Computational Science, mai 2025, vol. 3, n°5, pp. 427–439

[18]. Doumanas D., Soularidis A., Spiliotopoulos D., Vassilakis C., Kotis K., Fine-Tuning Large Language Models for Ontology Engineering: A Comparative Analysis of GPT-4 and Mistral, Applied Sciences, Février 2025, vol. 15, n°4, article 2146.

[19]. DeBellis M., Dutta N., Balaji A., Integrating Ontologies and Large Language Models to Implement Retrieval Augmented Generation, dans IOS Press (IAOA), janvier 2025, vol. 19

[20]. SPHN Consortium, Semantic Interoperability in Swiss Personalized Health Network, dans SPHN Technical Reports, avril 2024, rapport n°TR-2024-04, pp. 1–14.

[21]. Alarcón Moreno P., Wilkinson M. D., Clinical And Registry Entries Semantic Model (CARE SM): a knowledge graph model for patient data registries, dans CEUR Workshop Proceedings (SWAT4HCLS 2024), février 2024, vol 3890

[22]. Fédération Internationale du Diabète, Profil pays : République Démocratique du Congo, IDF Diabetes Atlas, 2024, 10e édition, p. 71-73.

[23]. Ministère de la Santé Publique (RDC), Normes et directives de la prise en charge du diabète sucré de type 2, Kinshasa : Programme National de Lutte contre le Diabète (PNLD), novembre 2013, 60 p.

[24] Atun R., Davies J. I., Gale E. A. M., Bärnighausen T., Diabetes in Sub-Saharan Africa: Challenges and opportunities, dans The Lancet Diabetes & Endocrinology, septembre 2014, vol. 2, n°9, pp. 675–684.

[25]. International Diabetes Federation (IDF), IDF Diabetes Atlas, 10e édition, Bruxelles, 2021, pp. 1–20.

[26]. Zhou Y., Shi Q., Chen X., Li S., Shen B., Ontologies applied in clinical decision support systems for diabetes, dans Journal of Sichuan University (Medical Sciences), 2023, vol. 54, n°1, pp. 208–216

[27]. Flory J.H., Ancker J.S., Kim S.Y.H., Kuperman G., Vickers A., Large language model GPT-4 compared to endocrinologist responses on initial choice of glucose-lowering medication under conditions of clinical uncertainty, dans Diabetes Care, février 2025, vol. 48, n°2, pp. 185–192

[28]. Toro S., Anagnostopoulos A.V., Bello S., Dynamic retrieval augmented generation of ontologies using artificial intelligence (DRAGON-AI), dans Journal of Biomedical Semantics, octobre 2024, vol. 15, n°1, art. 32, pp. 1–12

[29]. Seneviratne O., Das A.K., Chari S., Semantically enabling clinical decision support recommendations, dans Journal of Biomedical Semantics, juillet 2023, vol. 14, n°1, art. 32, pp. 1–10

[30]. Palojoki S., Semantic interoperability of electronic health records: systematic review, dans JMIR MedicalInformatics,octobre2024,vol.12,n°1,art.e53535,pp.1–14https://https:// medinform.jmir.org 2024/1/e53535

[31]. Nacanabo R., Debussche X., Rouamba M., Kamouni P., Mancini J., Kouanda S., Health literacy and health-related quality of life in type 2 diabetes: a cross-sectional study in Burkina Faso, in Diabetes Epidemiology and Management, 2021, vol. 3.

[32]. Ouedraogo I., Some B. M. J., Benedikter R., Diallo G., Health and eHealth Literacy of Patients With Diabetes in Low-Income Countries: Perspective From Guinea and Burkina Faso, dans JMIR Diabetes, décembre 2024, vol. 9

[33]. Abbasian M., Yousefi H., Elali M., Knowledge-infused LLM-powered conversational health agent: a case study for diabetes patients, dans arXiv preprint, février 2024, arXiv:2402.10153, 12 p.

[34]. K.Göğebakan., RUlu., A drug prescription recommendation system based on novel DIAKID ontology for T2DM and CKD patients, dans BMC Medical Informatics and Decision Making, Mars 2024, vol. 12, article 25, pp. 1–12

[35]. Ademola A., George C., Mapp G., Addressing the interoperability of electronic health records: the technical and semantic interoperability, preserving privacy and security framework, dans Applied System Innovation, Novembre 2024, vol. 7, n°6, article 116, pp. 1–11

[36]. Lippolis A. S., Mohammad Javad S., Ontology generation using large language models, dans arXiv preprint, mars 2025, arXiv:2503.05388.

[37]. Shimizu C., Hitzler P., Accelerating knowledge graph engineering with LLMs, dans arXiv preprint, novembre 2024, arXiv:2411.09601.

[38]. Noda R., Tanabe K., Ichikawa D., Shibagaki Y., GPT-4’s performance in supporting physician decision-making in nephrology multiple-choice questions, dans Scientific Reports, avril 2025, vol. 15, article 15439, pp. 1–9

[39]. Rangel J.C., de Farias T.M., Sima A.C., Kobayashi N., SPARQL generation: an analysis on fine-tuning OpenLLaMA for question answering over a life science knowledge graph, dans arXiv preprint, février 2024, arXiv:2402.04627, 15 p.

[40]. Dataquest, Using Excel with Python and Pandas, tutoriel en ligne, janvier 2024, Dataquest Blog.

[41]. Nebius,AI model performance metrics: in-depth guide, blog technique,février2024,Nebius Tech Blog

[42]. Amazon Web Services, Ground truth curation and metric interpretation best practices for evaluating generative AI question answering using FMEval, AWS ML Blog, mars 2024

[43]. Hamrouni B., Bourouis A., Korichi A., Brahmi M., Explainable ontology-based intelligent decision support system for business model design and sustainability, dans Sustainability, septembre 2021, vol. 13, n°17, article 9819, pp. 1–21

[44]. Lan Y., He G., Jiang J., Zhao W.X., Wen J.-R., Complex knowledge base question answering: a survey, dans arXiv preprint, août 2023, arXiv:2108.06688, 45 p.

[45]. Shen Y., Colloc J., Jacquet-Andrieu A., Guo Z., Constructing ontology-based cancer treatment decision support system with case-based reasoning, dans arXiv preprint, décembre 2018, arXiv:1812.01891, 15 p.

[46]. Mircha P., Joost D. W., Véronique L., Performance indicators on long-term care for older people in 43 high-income countries: literature review, web search and expert consultation, dans BMC Health Services Research, 2025, vol. 25, article 12573, pp. 1–16